pagenotfound-cli/cli.mjs

#!/usr/bin/env -S npx nodejsscript
/* jshint esversion: 11,-W097, -W040, module: true, node: true, expr: true, undef: true *//* global echo, $, pipe, s, fetch, cyclicLoop */
const url_drops= "https://pagenotfound.cz/drop/";
const { version, description, homepage }= s.cat("package.json").xargs(JSON.parse);
/**
 * @typedef {Object} Article
 * @property {string} title
 * @property {string} perex
 * @property {string} author
 * @property {string} loc
 * @property {string} drop
 * */
/**
 * @typedef {Object} Drop
 * @property {string} drop
 * @property {string} date
 * */
/**
 * @typedef {Object} Sitemap
 * @property {Article[]} articles
 * @property {Drop[]} drops
 * */
/**
 * @typedef {Object} State
 * @property {Sitemap} json
 * @property {string[]} changed Changed files
 * */
const pubDate= (function pubDateInner(){ // pubDate must be an RFC-822 date-time
	const intl= new Intl.DateTimeFormat("en-gb", {
		/* Wed, 02 Oct 2002 */ weekday: "short", day: "2-digit", month: "short", year: "numeric",
		/* 08:00:00 */ hour12: false, hour: "2-digit", minute: "2-digit", second: "2-digit",
		/* EST */ timeZoneName: "longOffset", timeZone: "UTC",
	});
	/** @param {string} date */
	return date=> {
		const D= new Date(date);
		const d= intl.format(D);
		// remove redundant second comma to make it RFC-822
		const red_comma_i= d.indexOf(",", d.indexOf(",")+1);
		return d.slice(0, red_comma_i) + d.slice(red_comma_i+1);
	};
})();

$.api()
.version(version)
.describe(description)
.command("pull", "Update article list")
.option("--git", "Update git repository")
.action(async function pull({ git: is_git= false }){
	if(is_git) s.run`git pull --rebase`;
	const { changed }= await sitemap().then(toRSS);
	echo("Changed files:", changed.length ? changed.join(", ") : "—");
	if(is_git) gitCommit(changed, "pull");
	$.exit(0);
})
.parse();

function gitCommit(files, des= "not specified"){
	if(!files.length || !s.run`git diff --numstat`.trim())
		return echo("Nothig todo");
	
	echo("Diff to save");
	s.run`git config user.name "Bot"`;
	s.run`git config user.email "${"zc.murtnec@naj.elrdna".split("").reverse().join("")}"`;
	s.run`git add ${files}`;
	s.run`git commit -m "Updated by bot – ${des}"`;
	s.run`git push`;
	s.run`git config --remove-section user`;
}
/**
 * @param {State} state
 * @returns {State} state
 * */
async function toRSS({ json, changed }){
	if(!changed.length) return { json, changed };
	
	const path= "rss.xml";
	const host= "https://pagenotfound.cz";
	const articles= json.articles.map(function({ title, perex, author, loc, drop }){
		return [
			"<item>",
			...[
				`<title>${title}</title>`,
				`<link>${host+loc}</link>`,
				`<guid>${host+loc}</guid>`,
				`<description>${perex}</description>`,
				`<dc:creator>${author}</dc:creator>`,
				`<pubDate>${pubDate(json.drops.find(d=> d.drop === drop).date)}</pubDate>`,
				`<category>${drop}</category>`,
			].map(l=> "\t"+l),
			"</item>"
		].map(l=> "\t"+l).join("\n");
	});

	const description = [
		"Page not found jsme založili z touhy po zábavné, kvalitní a inovativní žurnalistice.",
		"Chceme vám tu nabízet komplexní long ready, nečekané bonusy, multimediální obsah",
		"a hlavně texty, které nám samotným v ostatních médiích chybí. Budujte spolu s námi",
		"komunitu Page not found, dejte nám zpětnou vazbu na první drop, přihlaste se",
		"k odebírání newsletterů. Společně s vámi budeme moct naše cíle plnit rychleji."
	].join(" ");
	s.echo([
		`<?xml version="1.0" encoding="UTF-8" ?>`,
		`<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom">`,
		"<channel>",
		`	<title>Pagenotfound.cz</title>`,
		`	<link>${host}</link>`,
		`	<description>${description}</description>`,
		`	<atom:link href="${homepage}/raw/branch/main/rss.xml" rel="self" type="application/rss+xml" />`,
		...articles,
		"</channel>",
		"</rss>"
	].join("\n")).to(path);
	return { json, changed: [...changed, path] };
}
import { JSDOM } from "jsdom";
/** @returns {Promise<State>} */
async function sitemap(){
	const path=  "sitemap.json";
	/** @type {Sitemap} */
	const json= s.test("-f", path) ? s.cat(path).xargs(JSON.parse) : { drops: [], articles: [] };
	await syncDrops(json);
	const [ { drop: drop_last } ]= json.drops;
	
	const res= await fetch(url_drops+drop_last);
	if(res.status !== 200) return { json, changed: [] };
	
	const drop_articles= dropArticles(drop_last, json);
	const dom= new JSDOM(await res.text());
	const diff= [];
	for(const article of dom.window.document.querySelectorAll("article")){
		const loc= article.querySelector("a")?.href;
		if(!loc){
			echo("Article without link:", article.textContent);
			continue;
		}
		if(drop_articles.has(loc)) continue;
		diff.push({
			title: article.querySelector("h2").textContent.trim(),
			perex: article.querySelector("[class^=ArticleTile_perex]").textContent.trim(),
			author: article.querySelector("[class^=ArticleTile_author]").textContent.trim(),
			loc,
			drop: drop_last,
		});
	}
	if(!diff.length) return { json, changed: [] };

	json.articles.unshift(...diff);
	s.echo(JSON.stringify(json, null, "\t")).to(path);
	return { json, changed: [ path ] };
}
/**
 * Assumes that articles are sorted from newest to oldest
 * @param {Drop.drop} drop
 * @param {Sitemap} json
 * @returns {Set<Article.loc>}
 * */
function dropArticles(drop, { articles }){
	const out= new Set();
	for(const article of articles){
		if(article.drop !== drop) break;
		out.add(article.loc);
	}
	return out;
}
/** @param {Sitemap} json */
async function syncDrops(json){
	const [ { drop: drop_last } ]= json.drops;
	const i_index= drop_last.search(/\d/);
	const pre= drop_last.slice(0, i_index);
	const index= pipe(
		Number,
		i=> i+1,
		i=> i.toString().padStart(drop_last.length - i_index, "0"),
	)(drop_last.slice(i_index));

	const drop= pre+index;
	const res= await fetch(url_drops+drop, { method: "HEAD" });
	if(res.status !== 200) return json;

	const date= pipe(
		d=> new Date(d),
		d=> d.toISOString(),
	)(res.headers.get("date"));
	json.drops.unshift({ drop, date });
	return json;
}